Do you have Arc Fault turned on? The TIGO optimizers trigger the arc fault on my Sol-Arks. I monitor 3 Sol-Ark systems. The 2 with optimizers frequently trigger arc fault. The one with basic TIGO RSD does not have this problem.
Optimizers are SMPS, PWM pulsing into an inductor.
They will create conducted emissions with a fundamental and harmonics.
Given enough frequency-domain processing power, that would be distinguishable from the random noise of arcs.
SMPS need filtering to suppress EMI.
It is OK to put filters on AC wires of an inverter, but NOT on PV wires where that would "blind" AFCI to arcs.
I wonder if anybody has evaluated whether optimizers (with decent filtering of their noise) block arc-fault signals from reach AFCI, thereby defeating it? (The order of UL testing determines outcome?)
When AFCI and PWM are implemented in the same device, one approach would be to look for noise during the dead time between each switching. Remote device like optimizer doing the switching, that approach isn't possible. Dozens of optimizers, that many more randomly placed edges.
Outside interference (RF, etc.) could also be detected. AFCI needs to discriminate well enough.
I read that early AFCI implementations by SMA were having too much nuisance tripping, installers preferred other brands.
Sensitivity may be adjustable.